(RADIATOR) Problems with RADIUS accounting

[Alex Sharaz]

Hi,

I haven't given any more info re what's happening because I can't find any.

Trapeze says can't connect to radius server.
Our hp switches say the same thing with the added bit of info that if you
look at the logs on the switch you can see loads of retransmissions for
radius accounting.
I've tried changing the switch accounting retry time and (can't remember the
param nme) how long it will wait for a response from the radius server to no
avail.

In the production system the load balancer is performing L7 health checks on
all the radius servers and it doesn't have a problem

Switching on L4 tracing in radiator doesn't show anything else either. All
it shows is that everything is working fine. I can see auth and acct
requests being processed properly.

I'm fairly certain that the problem does lie with the response time from the
database. Hugh has suggested a tracing option that I'll try out to see
what's happening and go from there.

The annoying thing is that its happening for both db2 and mysql
Alex

On 17/04/2008 19:32, "Bjoern A. Zeeb" <bz-lists at cksoft.de> wrote:

> On Thu, 17 Apr 2008, Alex Sharaz wrote:
> 
> Hi,
> 
>> the same machine ... Which looked as if things might have worked. However, I
>> then pointed the Trapexe accounting at the devel server and almost
>> immediately started getting failure to connect to radius server messages on
>> the trapeze console. As it happened there was an error in an sql statement
>> for the sessionlog that dealt with updates. After I fixed this it looked as
>> if things were working o.k. The problem is that its now 5:32 on a Thursday
>> and there¹s not a lot of traffic around.
>> 
>> I really can¹t see anything wrong anywhere or why I¹m getting these errors.
> 
> While you are giving a fair amount of information on your hardware
> etc. you are not really saying what "these errors" are.
> 
> "Failure to connect to radius server" sounds strange as with UDP you
> cannot really connect but I read this as the Trapeze doesn't get a
> reply from the radius server for whatever request it was sending
> or some intermediate device sends a host/port unreachable or the like.
> 
> In this case a Trace 4 log on the radius server could tell you if it
> is seeing the request coming in, if it is able to handle it
> correctly and if it sends out a reply. If this is all true and the
> Trapeze still complains, maybe tcpdump on the radius server (and if
> possible at the Trapeze side - the other side of the load balancer)
> could be interesting as well to see if you lose the packets somewhere.
> 
> If you think it's a problem with Radiator talking to the DB a Trace 4
> should really tell you about problems as well.
> 
> 
> Regards,
> Bjoern
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <http://www.open.com.au/pipermail/radiator/attachments/20080418/d2cac2c7/attachment.ksh>